菜单
Generate scored A/B test variations of headlines, hooks, CTAs, and intros ranked by optimization goal.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Process multiple content pieces through a prioritized, checkpointed queue with progress tracking and per-piece quality gates
Add a custom MCP connector — connect any API or service to ContentForge via .mcp.json configuration.
Track content quality scores, pipeline timing, and compliance trends with insights and alerts.
Audit content library for freshness decay, coverage gaps, and optimization opportunities.
Generate research-backed content briefs with keywords, competitors, intent, and SEO strategy from a topic.
Plan content calendars with scheduling, deadlines, team assignments, and Google Calendar sync.
| name | cf-variants |
| description | Generate scored A/B test variations of headlines, hooks, CTAs, and intros ranked by optimization goal. |
| argument-hint | [content-element] |
| effort | medium |
Generate multiple scored variations of any content element — headlines, hooks, CTAs, introductions, or conclusions — ranked by composite quality score and aligned to your optimization goal.
Use /contentforge:cf-variants when you need:
For full content production, use /contentforge:create-content instead.
For refreshing entire pieces, use /contentforge:content-refresh instead.
Takes a single content element and generates N variations (3-10), scores each across 6 quality dimensions, ranks by composite score weighted toward your optimization goal, and presents the top 3 with detailed reasoning and A/B test setup guidance.
Process Flow:
Minimum Required:
headline | hook | cta | intro | conclusionOptional:
clicks | engagement | conversions | readability (default: engagement)/contentforge:brand-setup)/contentforge:cf-variants
Prompts you for:
/contentforge:cf-variants "AI Will Transform Your Business in 2026" --type=headline --count=5 --goal=clicks --brand=AcmeMed
/contentforge:cf-variants --source-doc=https://docs.google.com/document/d/XYZ123 --element=headline --count=7 --goal=conversions
Extracts the headline from the document and generates 7 conversion-optimized alternatives.
Parse the source element and assess its current performance profile:
Analysis Output:
Source Element Analysis
-------------------------------------------------------------
Element: "AI Will Transform Your Business in 2026"
Type: Headline
Word Count: 8
Structure: Statement (declarative)
Emotional Register: Aspirational (moderate)
Specificity: Low (no numbers, no concrete outcome)
Curiosity Gap: Low (no open question, no surprise)
Keyword Presence: "AI" present, "business transformation" absent
Brand Voice: Neutral (no brand-specific markers)
Improvement Opportunities:
- Add specificity (numbers, concrete outcomes)
- Increase curiosity gap (why 2026? what changes?)
- Strengthen emotional pull (urgency or fear of missing out)
- Include target keyword for SEO
-------------------------------------------------------------
Generate N variations using two complementary patterns:
Humanizer Patterns Applied:
Reviewer Logic Applied:
Variation Strategies by Element Type:
| Element Type | Strategy Mix |
|---|---|
| Headline | Question, number-driven, how-to, contrarian, emotional, keyword-rich |
| Hook | Statistic-lead, story-lead, question-lead, bold-claim, problem-agitate |
| CTA | Action-verb, benefit-driven, urgency, social-proof, low-friction |
| Intro | Problem-first, story-first, data-first, contrarian-first, question-first |
| Conclusion | Summary-CTA, future-vision, callback, challenge, resource-link |
Each variation is scored across 6 dimensions on a 0-10 scale:
| Dimension | Weight (Default) | What It Measures |
|---|---|---|
| Clarity | 0.20 | Instantly understandable, no ambiguity |
| Emotional Appeal | 0.20 | Triggers feeling (curiosity, urgency, aspiration) |
| Specificity | 0.20 | Concrete details, numbers, outcomes |
| Curiosity Gap | 0.15 | Creates desire to read more |
| Keyword Presence | 0.10 | SEO keyword naturally included |
| Brand Voice Fit | 0.15 | Matches registered brand personality |
Goal-Based Weight Adjustments:
When an optimization goal is specified, dimension weights shift:
| Dimension | Default | Clicks | Engagement | Conversions | Readability |
|---|---|---|---|---|---|
| Clarity | 0.20 | 0.15 | 0.15 | 0.20 | 0.35 |
| Emotional Appeal | 0.20 | 0.25 | 0.30 | 0.25 | 0.10 |
| Specificity | 0.20 | 0.20 | 0.15 | 0.25 | 0.15 |
| Curiosity Gap | 0.15 | 0.25 | 0.20 | 0.10 | 0.10 |
| Keyword Presence | 0.10 | 0.05 | 0.05 | 0.05 | 0.10 |
| Brand Voice Fit | 0.15 | 0.10 | 0.15 | 0.15 | 0.20 |
Sort variations by composite score (weighted sum), present top 3 with reasoning.
Provide actionable test parameters:
SYNTHETIC EXAMPLE — fabricated for illustration. All headlines, statistics, and scores below are invented; never reuse the invented statistics in real output.
================================================================
CONTENTFORGE VARIANT SCORECARD
================================================================
Source: "AI Will Transform Your Business in 2026"
Element Type: Headline
Optimization Goal: Clicks
Variations Generated: 5
Brand: AcmeMed (Authoritative, Professional)
================================================================
VARIATION SCORES (Ranked by Composite)
----------------------------------------------------------------
#1 "The 5 AI Shifts That Will Reshape Healthcare
Revenue by 2027"
┌────────────────────────────────────────────────┐
│ Clarity: 8.5 ████████░░ │
│ Emotional Appeal: 8.0 ████████░░ │
│ Specificity: 9.5 █████████░ │
│ Curiosity Gap: 9.0 █████████░ │
│ Keyword Presence: 7.0 ███████░░░ │
│ Brand Voice Fit: 8.5 ████████░░ │
├────────────────────────────────────────────────┤
│ COMPOSITE SCORE: 8.6 / 10 │
└────────────────────────────────────────────────┘
Why it works: Number-driven ("5 shifts") creates
scannable promise. "Reshape revenue" adds concrete
stakes. Industry-specific ("healthcare") signals
relevance. Curiosity gap: which 5 shifts?
#2 "Why 73% of Executives Are Wrong About AI
— And What It Will Cost Them"
┌────────────────────────────────────────────────┐
│ Clarity: 7.5 ███████░░░ │
│ Emotional Appeal: 9.0 █████████░ │
│ Specificity: 8.5 ████████░░ │
│ Curiosity Gap: 9.5 █████████░ │
│ Keyword Presence: 6.5 ██████░░░░ │
│ Brand Voice Fit: 7.5 ███████░░░ │
├────────────────────────────────────────────────┤
│ COMPOSITE SCORE: 8.3 / 10 │
└────────────────────────────────────────────────┘
Why it works: Contrarian angle ("wrong about AI")
provokes click. Specific stat (73%) adds authority.
Fear element ("what it will cost") drives urgency.
Strong curiosity gap: wrong about what, exactly?
#3 "How AI-Powered Diagnostics Cut Time-to-Treatment
by 40% (2026 Data)"
┌────────────────────────────────────────────────┐
│ Clarity: 9.0 █████████░ │
│ Emotional Appeal: 7.0 ███████░░░ │
│ Specificity: 9.5 █████████░ │
│ Curiosity Gap: 7.5 ███████░░░ │
│ Keyword Presence: 9.0 █████████░ │
│ Brand Voice Fit: 8.5 ████████░░ │
├────────────────────────────────────────────────┤
│ COMPOSITE SCORE: 8.1 / 10 │
└────────────────────────────────────────────────┘
Why it works: How-to structure sets clear
expectation. Concrete outcome ("40%") adds
credibility. Year tag ("2026 Data") signals
freshness. Strong SEO keyword placement.
#4 "AI in Healthcare Is No Longer Optional
— Here's Your 90-Day Action Plan"
┌────────────────────────────────────────────────┐
│ Clarity: 8.5 ████████░░ │
│ Emotional Appeal: 7.5 ███████░░░ │
│ Specificity: 7.0 ███████░░░ │
│ Curiosity Gap: 7.0 ███████░░░ │
│ Keyword Presence: 8.0 ████████░░ │
│ Brand Voice Fit: 8.0 ████████░░ │
├────────────────────────────────────────────────┤
│ COMPOSITE SCORE: 7.6 / 10 │
└────────────────────────────────────────────────┘
Why it works: Urgency framing ("no longer
optional"). Actionable promise ("90-day plan")
reduces perceived effort. Good keyword placement.
#5 "What Happened When 12 Hospitals Deployed AI
Last Quarter"
┌────────────────────────────────────────────────┐
│ Clarity: 8.0 ████████░░ │
│ Emotional Appeal: 7.0 ███████░░░ │
│ Specificity: 8.0 ████████░░ │
│ Curiosity Gap: 8.5 ████████░░ │
│ Keyword Presence: 6.0 ██████░░░░ │
│ Brand Voice Fit: 7.5 ███████░░░ │
├────────────────────────────────────────────────┤
│ COMPOSITE SCORE: 7.4 / 10 │
└────────────────────────────────────────────────┘
Why it works: Story-lead ("what happened when")
creates narrative pull. Specific ("12 hospitals",
"last quarter") builds trust. Curiosity gap is
strong but keyword integration is weaker.
================================================================
TOP 3 RECOMMENDATIONS
================================================================
1. BEST FOR CLICKS: Variant #1
"The 5 AI Shifts That Will Reshape Healthcare
Revenue by 2027"
Rationale: Number-driven headlines consistently
outperform in CTR testing. The specificity of
"5 shifts" combined with revenue stakes creates
both scannability and urgency.
2. BEST FOR ENGAGEMENT: Variant #2
"Why 73% of Executives Are Wrong About AI
— And What It Will Cost Them"
Rationale: Contrarian angles drive shares and
comments. The implicit challenge ("are you in
the 73%?") creates personal relevance that
extends dwell time and social sharing.
3. BEST FOR SEO: Variant #3
"How AI-Powered Diagnostics Cut Time-to-Treatment
by 40% (2026 Data)"
Rationale: Natural keyword integration with
"AI-powered diagnostics" matches high-intent
search queries. The "how" structure aligns with
featured snippet patterns.
================================================================
A/B TEST SETUP GUIDANCE
================================================================
Recommended Test Configuration:
- Variants to Test: #1 vs. #2 (or #1 vs. original)
- Sample Size: 1,000+ impressions per variant
- Duration: 7-14 days (avoid holidays, weekends-only)
- Primary Metric: Click-through rate (CTR)
- Secondary Metrics: Time on page, scroll depth,
bounce rate
- Confidence Level: 95% statistical significance
- Winner Criteria: >10% relative CTR improvement
Test Channels:
- Email subject lines: Send to equal-sized segments
- Blog headlines: Use CMS A/B testing plugin
- Social posts: Run as separate posts, same time
- Ad copy: Split test in Google/Meta Ads Manager
Fallback Plan:
If no variant achieves statistical significance
after 14 days, deploy Variant #1 (highest composite
score) as the safe default.
================================================================
How quickly a reader grasps the meaning. Penalizes jargon, ambiguity, overly long constructions, and unclear referents. Rewards plain language, active voice, and immediate comprehension.
The strength of the emotional trigger. Measures curiosity, urgency, aspiration, fear, empathy, or challenge. A score of 5 is emotionally neutral; 8+ means strong pull. Penalizes flat, corporate-speak phrasing.
Concrete details vs. vague promises. Numbers, percentages, timeframes, named entities, and measurable outcomes score high. Generic phrases like "improve your results" score low.
The degree to which the element creates a desire to read more. Questions, incomplete information, surprising claims, and "what happened next" structures score high. Fully self-contained statements score lower.
Whether the primary SEO keyword appears naturally. Scores high for natural inclusion in prominent position. Scores low for absence, forced placement, or keyword stuffing. Evaluated only if a keyword is provided or extractable from brand profile.
Alignment with the registered brand personality. If no brand is specified, scores based on professional consistency. Checks tone (authoritative vs. casual), terminology (approved vs. banned words), and guardrails (topics to avoid).
Before Variants:
/contentforge:create-content — Generate the base content first/contentforge:content-refresh — Update content before variant testingAfter Variants:
/contentforge:cf-analyticsThis skill uses existing Humanizer and Reviewer agent patterns — no additional agents required.
Agents: Humanizer (patterns), Reviewer (scoring logic) Output: Scored variation cards with top 3 recommendations + A/B test setup guidance